Decision making

Decision Making
Despite the errors that occur when we make judgment, these
judgment form an important part of the database for the process of
decision making. Decision making includes a choice between
alternatives. With the increase in the number of alternatives
available the probability/chance of alternatives being wrong also
increases – leading to the increase of risk/uncertainty in choice
According to the threshold approach of choice (Clemen,1991), if a
decision depends on the likelihood of another event happening,
then the attractiveness of the option should increase as the
probability of the other event increases. Once that probability
reaches a minimum level of certainty, the alternative would be
chosen.
Decisions which involve over-confidence in judgment attain
the minimum level of certainty too easily leading to choice of
wrong/un-rewarding choices

Does the minimum level of certainty vary with culture?

Expected Utility: A Normative Approach
Economists are interested in the factors involved in choice
and what type of model describes rational choice behavior.
One of the well-established theories of decision making is
expected utility theory
the theory states that when faced with some type of uncertain
choice, we make our decisions based on two factors
1) the expected utility of the outcomes
2) their respective probability
Utility refers to whatever end a person would like to achieve, be it
happiness, money or something else. Broome (1991) suggests –
utility refers to the amount of good that comes out of a decision.
Thus while making decisions we weigh the good that might come
out of each alternative against the cost of that alternative. We also
access the probability of each alternative occurring. Whatever
alternative provides the best combination of “good” and
“likelihood” will be the chosen one. Consider
flip a coin; if it turns up head, you get $40
Roll a dice, if it come up 4, you get $50.
Which option would you chose?
Violation of Expected Utility
One of the normative predictions made by expected utility theory is
that our choices should show invariance's; that is, a decision
maker’s choice should not depend on the way a choice is presented.
If I prefer choice A over choice B in situation 1, then I should prefer
choice A over choice B in situation 14 (as long as A and B are
identical in the two situations)
People often switch their preferences of one outcome over another,
based on how these outcomes are presented, demonstrating
irrationality. Consider the preference reversal shown (Lichtenstein &
Slovic, 1971). Their general procedure involved having subjects look
at two different gambles and decide – (1) which gamble they would
like to play & (2) how much the gamble was worth.
1) 80% chance to win $ 4.00 4) 10% chance to win $ 40.00
20% chance to lose $ 0.50 90% chance to lose $ 1.00

2) 95% chance to win $ 3.00 5) 50% chance to win $ 6.50

5% chance to lose $ 2.00 50% chance to lose $ 1.00

3) 99% chance to win $ 4.00 6) 33% chance to win $ 16.00

1% chance to lose $ 1.00 67% chance to lose $ 2.00
Lichtenstein & Slovic expected that the choice of which gamble
to play would be influenced by the probability of winning,
whereas the choice of the selling price for the gamble would
depend on the potential dollar amount to be won. Why is this
preference reversal irrational?
The preference reversal phenomena demonstrates the
inadequacy of expected utility as a descriptive model of
decision making. The expected utility model fails to provide a
good description of how we make choices in many
circumstances because it assumes too much; humans rarely
have all the information necessary to make a decision. Even if
they did, they lack the ability to combine and weigh the
information accurately
Multiattribute Utility Theory
(MAUT)
• What if the choices differ on many
dimensions?
• Example: Choosing a major
• Majors differ in many ways: your interest in
them, the job market after graduation, the
faculty, etc.
• How should one choose?
Steps in MAUT
1. Break the decision down into its important dimensions.
2. Determine the relative weight (importance) of each
dimension.
3. List all of the alternatives.
4. Rank the alternatives along each dimension.
5. Multiply each ranking by the appropriate weight.
6. Choose the alternative with the highest value.
An example of MAUT in action
Prospect Theory
One popular alternative to expected utility theory is
Kahneman & Tverskys (1979) prospect theory. Prospect theory is a
descriptive model of decision making that attempts to describe how
we make decisions and why our decisions violate the expected
utility model. The theory states
decisions are not valued based on the absolute value of the
end result, as proposed by expected utility; instead we value
decisions based on the amount of gain or loss from what we have
right now. It also adds that gains and losses are on different scales
of value.
The value we attach to gain increases
more slowly as a function of the size of
the gains than does the (negative) value
we place on the loses as a function of
the size of the loss. Basically we feel
losses more acutely then we feel gains;
the psychological pain associated with
losing $50 is greater than the
psychological pleasure of gaining $50.
prospect theory predicts that people will
be especially aversive to loss and will
show difference in preference
depending on how alternatives are
presented or framed
Framing – is the term used to describe the effects on our
decisions oh how a scenario is presented.
Prospect theory predicts our preferences will change
whenever our reference point changes. Decisions can be
influenced by how information is presented. If information is
presented in terms of a positive “gain frame”, we will be more
likely to avoid risk (risk averse) and pick a sure bet. However if
the same information is presented in a negative “loss frame”,
we will be more likely to take a risk (risk prone) to avoid loses.
Consider the results of a classic study by Tversky &
Kahneman (1981). Subjects were presented with this scenario
and two choices
Imagine that the U.S. is preparing for the outbreak of an unusual Asian
disease, which is expected to kill 600 people. Two alternative programs
for combating the disease have been proposed. Assume that the exact
scientific estimate of the consequences of the program is as follows:
If program A is adopted, 200 people will be saved.
If program B is adopted, there is a 1/3 probability that 600 people will be
saved and 2/3 probability that no one will be saved.
Other subjects were presented with exactly the same problem but with
different choices
If program C is adopted, 400 people will die
If program D is adopted, there is 1/3 probability that nobody will die, &
2/3 probability that 600 people will die.
Psychological Accounting
this principle states that people will make different decisions
depending on how the outcomes is felt or perceived. Consider
1) imagine you have decided to see a play for which admission is $10 a
ticket. As you enter the theater, you discover that you have lost a $10
bill. Would you still pay $10 for a ticket to the play?
2) Imagine that you have decided to see a play for which admission is
$10 a ticket. As you enter the theater, you discover that you have
lost the ticket. The seat was not marked and the ticket cannot be
recovered. Would you pay $10 for a ticket to the play?
In Kahneman & Tversky (1981) original study the subjects were less willing
to purchase a ticket in scenario 2. Why could this be?
Sunk Cost
the sunk cost effect is another interesting variation of the notion
of psychological accounting. This effect was demonstrated by
Arkes and Blumer (1985) In one experiment
subjects were to imagine that they had purchased tickets for two
different ski trips: one ticket (for trip to Wisconsin) cost $50,
while the other ticket (for trip to Michigan) cost $100. the
scenario made it clear that the trip to Wisconsin was preferable
because it would be more enjoyable.
then a complication arose: the two trips were on the same
weekend and the tickets were non refundable. Which trip would
you choose to go on?
Affect and Decision Making
Positive and negative outcomes feel different to us, with
predictable implications for the decisions we make. Affect thus is an
important determinant of decision making, and can have sizable
impact on psychological accounting process.
Hsee & Rottenstreich (2004) make this point by highlighting an
important dimension of choice that interacts with affect, which they
term scope; it basically refers to the sweep of a decision or action –
how much impact will it have? Consider
suppose you gave $10 to help save one endangered tiger---
feels good. Now much would you give to save 4 endangered tigers?
The answer depends on whether the subjective value you derive from
saving tigers is somehow multiplicative?
The authors propose a dual-process view of the relative impact of
scope and subjective value on decision making. Their dual processes
are – a deliberate mode (which would map into the conscious
reasoning) and an affective mode (which would map onto the
unconscious reasoning).
when we’re in a deliberate decision making mode, we value things by
calculation (4 > 2); while in an affective decision making mode we
value things by feeling (help tigers). In deliberate decision making
mode as scope increases subjective value increases correspondingly,
while in affective decision making mode scope doesn’t matter nearly
as much and we are affected by the presence/absence of a stimulus